That the history of our species came in stages was an idea that came in stages. Aristotle saw the formation of political entities as a tripartite process: first we had families; next we had the villages into which they banded; and finally, in the coalescence of those villages, we got a governed society, the polis. Natural law theorists later offered fable-like notions of how politics arose from the state of nature, culminating in Thomas Hobbes’s mid-seventeenth-century account of how the sovereign rescued prepolitical man from a ceaseless war of all against all.

But it was Jean-Jacques Rousseau, a hundred years later, who popularized the idea that we could peer at our prehistory and discern developmental stages marked by shifts in technology and social arrangements. In his Discourse on the Origin and the Foundation of Inequality (1755), humans went from being solitary brutes to companionable, egalitarian hunter-gatherers; but with the rise of metallurgy and agriculture, things had taken a dire turn: people were civilized, and humanity was ruined. Once you found yourself cultivating a piece of land, ownership emerged: the field you toiled over was yours. Private property led to capital accumulation, disparities of wealth, violence, subjugation, slavery. In short order, political societies “multiplied and spread over the face of the earth,” Rousseau wrote, “till hardly a corner of the world was left in which a man could escape the yoke.”

Even people who rejected his politics were captivated by his origin story. In the nineteenth century, greater empirical rigor was brought to the conjectural history that Rousseau had unfolded. A Danish archaeologist partitioned prehistory into the Stone, Bronze, and Iron Ages; a British one split the Stone Age into the Paleolithic and the Neolithic. For the emerging discipline of anthropology, the crucial stages were set out in Ancient Society (1877) by the American ethnologist Lewis Henry Morgan. Human beings, he concluded, had emerged from a hunter-gatherer phase of “savagery” to a sedentary “barbarian” era of agriculture, marked by the domestication of cereal grains and livestock. Technologies of agriculture advanced, writing arose, governed towns and cities coalesced, and civilization established itself. Morgan’s model of social evolution, presaged by Rousseau, became the common understanding of how political society came about.

Then came another important stage in the story of stages. In the 1930s, the Australian archaeologist V. Gordon Childe synthesized the anthropological and archaeological findings of his predecessors: after a Paleolithic era of hunting and gathering in small bands, a Neolithic revolution saw the rise of agriculture (again, mainly harvesting cereals and herding ruminants), a soaring population, sedentism, and finally what he called the “urban revolution,” distinguished by large, dense settlements, administrative complexity, public works, hierarchy, systems of writing, and states. This basic story of social evolution has been refined and revised by later scholarship. (One recent point of emphasis is that grain, being storable and hard to hide, lent itself to taxation.) But it’s mainly taken to be—as we like to say these days—directionally correct. There was a stepwise connection, we think, between sowing cereals in our primeval past and waiting in line at the Department of Motor Vehicles.

The Dawn of Everything: A New History of Humanity, by the anthropologist David Graeber and the archaeologist David Wengrow, assails the proposition that there’s some cereals-to-states arrow of history. A mode of production, they insist, doesn’t come with a predetermined politics. Societies of hunter-gatherers could be miserably hierarchical; some indigenous American groups, fattened on foraging and fishing, had vainglorious aristocrats, patronage relationships, and slavery. Agriculturalist communities could be marvelously democratic. Societies could have big public works without farming. And cities—this is a critical point for Graeber and Wengrow—could function perfectly well without bosses and administrators.

They eloquently caution against assuming that what actually happened had to happen: in particular, that once human beings came up with agriculture, their descendants were bound to be subjects of the state. We’ve been persuaded that large-scale communities require some people to give orders and others to follow them. But that’s only because states, smothering the globe like an airborne toxic event, have promoted themselves as a natural and inevitable development. (Graeber, who died last year at fifty-nine, was, among other things, an anarchist theorist, advocate, and activist.)

Both Hobbes and Rousseau, The Dawn of Everything argues, have led us badly astray. Now, if you don’t like states, you’ll naturally be rankled by the neo-Hobbesian claim, made in books like Steven Pinker’s The Better Angels of Our Nature (2011), that we’re ennobled—less prone to violence and generally nicer—when we submit to centralized authority, with its endless rules and bureaucratic strictures. Yet the neo-Rousseauians aren’t a big improvement, in Graeber and Wengrow’s account, for they represent the sin of despair. It’s all very well for the anthropologist Marshall Sahlins (Graeber’s Ph.D. adviser at the University of Chicago) to have talked about the superiority—moral and hedonic—of the hunting-foraging lifestyle to our own, or for Jared Diamond to have said, for similar reasons, that the agricultural revolution was a terrible mistake. The fact remains that our planet can’t sustain 7.7 billion people with hunting and foraging: there’s no going back when it comes to the rise of agriculture. If growing grain leads to governments—if it entails our submission to state power—there’s nothing much to be done; we are left to watch YouTube videos of happy !Kung people in the Kalahari and sigh over our mug of fair trade coffee.


Graeber and Wengrow reject such paralytic pessimism. They believe that social evolutionism is a con, aimed at making us think that we had no choice but to forfeit our freedom for food and that the states we find everywhere are the inexorable result of developments ten or twelve thousand years ago. The Hobbesian spin leads to pigheaded triumphalism; the Rousseauian spin leads to plaintive defeatism. In their view we should give up both and reject the inevitability of states. Maybe history doesn’t supply any edifying counterexamples—lasting, large-scale, self-governing, nondominating communities sustained by mutual aid and social equality. But, Graeber and Wengrow argue, our prehistory does. To imagine a future where we are truly free, they suggest, we need to grasp the reality of our Neolithic past—to see what nearly was ours.

The Dawn of Everything, chockablock with archaeological and ethnographic minutiae, is an oddly gripping read. Graeber, who did his fieldwork in Madagascar, was well known for his caustic wit and energetic prose, and Wengrow, too, has established himself not only as an accomplished archaeologist working in the Middle East but as a gifted and lively writer. A volume of macrohistory—even of anti-macrohistory—needs to land its points with some regularity, and Graeber and Wengrow aren’t averse to repeating themselves. But for the most part they convey a sense of stakes and even suspense. This is prose it’s easy to surrender to.

But should we surrender to its arguments? One question is how persuasive we find the book’s intellectual history, which mainly unspools from the early Enlightenment to the macrohistorians of today and tells of how consequential truths about alternate social arrangements got hidden from view. Another is how persuasive we find the book’s prehistory, in particular its parade of large-scale Neolithic communities that, Graeber and Wengrow suspect, were self-governing and nondominating. On both time scales, The Dawn of Everything is gleefully provocative.

Among its most arresting claims is that European intellectuals had no concept of social inequality before the seventeenth century because the concept was, effectively, a New World import. Indigenous voices, particularly from the Eastern Woodlands of North America, helped enlighten the Enlightenment thinkers. Graeber and Wengrow focus on a dialogue that the Baron de Lahontan, who had served with the French army in North America, published in 1703, ostensibly reproducing conversations he had during his New World sojourn with a Wendat (Huron) interlocutor he named “Adario,” based on a splendid Wendat statesman known as Kandiaronk. Graeber and Wengrow say that Kandiaronk, in his opposition to dogma, domination, and inequality, embodied what they call “the indigenous critique.” And it was immensely powerful: “For European audiences, the indigenous critique would come as a shock to the system, revealing possibilities for human emancipation that, once disclosed, could hardly be ignored.”

Mainstream historians, such as Richard White, seem inclined to think that Adario’s voice is partly Kandiaronk’s and partly Lahontan’s. Graeber and Wengrow, by contrast, maintain that (allowing for embellishment) Adario and Kandiaronk were one and the same. It’s of no consequence, they say, that Adario’s claims that his people had no concept of property, no inequality, and no laws were (as they acknowledge) simply not true of the Wendat. Nor is any weight given to the fact that Adario shares Lahontan’s anticlerical Deism, expresses specific critiques of Christian theology associated with Pierre Bayle and other early philosophes, and offers a strikingly detailed critique of the abuses of the French judiciary. If the dialogue presents no conceptually novel arguments, that’s to be expected; after all, Graeber and Wengrow say, “there are only so many logical arguments one can make, and intelligent people in similar circumstances will come up with similar rhetorical approaches.” Maybe so. Still, our understanding of the indigenous critique would have been strengthened had they tried to determine what, for its time, was and was not distinctive in this dialogue.

But then they would have had to discard the thesis that Europeans, before the Enlightenment, lacked the concept of social inequality. This claim is plainly wide of the mark. Look south, and you find that Francisco de Vitoria (circa 1486–1546), like others of the School of Salamanca, had much to say about social inequality; and he, in turn, could cite eminences like Gregory the Great, who in the sixth century insisted that all men were by nature equal, and that “to wish to be feared by an equal is to lord it over others, contrary to the natural order.” Look north, and you find the German radical Thomas Müntzer in 1525 spurring on the Great Peasants’ Revolt:


Help us in any way you can, with men and with cannon, so that we can carry out the commands of God himself in Ezekiel 14, where he says: “I will rescue you from those who lord it over you in a tyrannous way…”

A vehement opposition to domination and to social inequality was certainly part of the Radical Reformation. Consider the theory and practice, in the same period, of such Anabaptist groups as the Hutterites, among whom private property was replaced by the “community of goods” and positions of authority subject to election.

Curiously, Graeber and Wengrow even hurry past the famous Montaigne essay from 1580 that takes up an episode in which explorers brought three Tupinamba from South America to the French court. The Tupinamba marveled that people at court should defer to the diminutive King Charles IX rather than to someone they selected out of their own ranks. They further marveled, Montaigne writes, that “there were amongst us men full and crammed with all manner of commodities” while others “were begging at their doors, lean and half-starved with hunger and poverty.” The Tupinamba wondered that these unfortunates “were to suffer so great an inequality and injustice, and that they did not take the others by the throats, or set fire to their houses.” It’s as if Graeber and Wengrow feared that this indigenous critique would detract from the shock to the system they associate with Kandiaronk.

What about their positioning of Rousseau? Following Émile Durkheim and others, they insist that his how-things-turned-bad story was never meant literally; it was merely a thought experiment. It’s true that the Discourse has a sentence to that effect: “One must not take the kind of research which we enter into as the pursuit of truths of history, but solely as hypothetical and conditional reasonings, better fitted to clarify the nature of things than to expose their actual origin.” But more plausible interpretations—notably the one offered by the intellectual historian A.O. Lovejoy—take that disclaimer to be a publishing precaution, or what Lovejoy calls “the usual lightning-rod against ecclesiastical thunderbolts.” The account is simply too detailed (metallurgy, Rousseau hypothesizes, arose from observing volcanic lava) to think he wasn’t serious about it.

Then Graeber and Wengrow repeat the familiar line that Rousseau thought everything was great until the state arose, while Hobbes thought everything was rotten. That’s why they say that Rousseau’s version of human history, just as much as Hobbes’s, has “dire political implications”—if granaries inevitably mean governments, “the best we can hope for is to adjust the size of the boot that will forever be stomping on our faces.”

Yet those implications don’t follow. In fact, Graeber and Wengrow have read past the fact that Rousseau and Hobbes were, on a critical point, in agreement: in the period that directly preceded the rise of the state, things were awful. Where Hobbes talked about a bellum omnium contra omnes, Rousseau invoked a “black inclination to harm one another.” You could say that Rousseau starts his story earlier than Hobbes (Lovejoy attentively counted four stages that come before political society in the Discourse, though you could draw the lines slightly differently); but the two wind up in the same place. The problem Rousseau identified is that the wealthy sold us on a rigged social compact that secured their interests at the expense of our freedom. And the solution wasn’t to return to the happy days of foraging and hunting; it was to craft a better social compact.

Graeber and Wengrow’s most significant claim, in the realm of intellectual history, is that “our standard historical meta-narrative about the ambivalent progress of human civilization” was “invented largely for the purpose of neutralizing the threat of indigenous critique”—that those grain-to-government stages represented a “conservative backlash” against the voices of freedom. They were designed to persuade us that we can’t do without obedience to centralized authority and should bloody well do as we’re told. Let’s put aside the perplexing inference that Rousseau’s Discourse on Inequality, in this account, at once promulgated the indigenous critique and smothered it. When we look at prominent social evolutionists, do we find apologists for centralized authority?

Rather the opposite. “Centralization is the tendency and the result of the institutions of arbitrary and despotic governments,” Lewis Henry Morgan maintained in his 1852 lecture “Diffusion Against Centralization,” denouncing a political order in which “property is the end and aim.” In Ancient Society, he aimed to revitalize, not neutralize, a politics of emancipation. “Democracy in government, brotherhood in society, equality in rights and privileges, and universal education, foreshadow the next higher plane of society,” he wrote.

Morgan’s tripartite scheme was revisited in Thorstein Veblen’s The Theory of the Leisure Class (1899). Peaceable and productive savages, in Veblen’s telling, gave way to more predatory and less productive barbarians; the rise of property rights and state power is essentially an outgrowth of patriarchy. But Veblen was hostile to determinism of the sort he found in Marx. What he favored was not surrender to the status quo but a nonstatist version of socialism, which some scholars have labeled anarchism. V. Gordon Childe, for his part, was a socialist with syndicalist tendencies who had hopes for radically different political arrangements.

In the mid-twentieth century, when social evolutionism fell from favor among anthropologists, its most vigorous advocate in the discipline was Leslie A. White. And White—who trained Sahlins, who trained Graeber—was a socialist leery of statism. Perhaps the most notable recent rendering of the cereals-to-states story appears, with novel elaborations, in Against the Grain by James C. Scott, who’s also the author of Two Cheers for Anarchism. If this metanarrative was purpose-built to reconcile us to an impoverished status quo, it’s curious that its greatest exponents advocated political transformation.

Graeber and Wengrow could be all wrong in their intellectual history, of course, and completely right about our Neolithic past. Yet their mode of argument leans heavily on a few rhetorical strategies. One is the bifurcation fallacy, in which we are presented with a false choice of two mutually exclusive alternatives. (Either Adario is Kandiaronk or Kandiaronk has no presence in Lahontan’s dialogue.) Another is what’s sometimes called the “fallacy fallacy”: because a bad argument is made for a conclusion, the conclusion must be false; or because a bad argument has been made against a conclusion, the conclusion must be true. And the absence of evidence routinely serves as evidence of absence. Through a curious rhetorical alchemy, the argument that a claim isn’t impossible gets transmuted into an argument that the claim is true.

Graeber and Wengrow tend to introduce a conjecture with the requisite qualifications, which then fall away, like scaffolding once a building has been erected. Discussing the Mesopotamian settlement of Uruk, they caution that anything said about its governance is speculation—we can only say that it didn’t have monarchy. The absence of a royal court is consistent with all sorts of political arrangements, including rule by a bevy of high-powered families, by a managerial or military or priestly elite, by ward bosses and shifting council heads, and so on. Yet a hundred pages later, the bifurcation fallacy takes effect—there’s either a royal boss or no bosses—and we’re assured that Uruk enjoyed “at least seven centuries of collective self-rule.” A naked “what if?” conjecture has wandered off and returned in the three-piece suit of an established fact.

A similar latitude is indulged when we visit the Trypillia Megasites (4100–3300 BC) in the forest-steppe of Ukraine. The largest of these settlement areas, Taljanky, is spread over 1.3 square miles, archaeologists have discovered more than a thousand houses there, and Graeber and Wengrow tell us that the per-site population was, in some cases, probably well over 10,000 residents. “Why would we hesitate to dignify such a place with the name of ‘city’?” they ask. Because they see no evidence of centralized administration, they declare it to be “proof that highly egalitarian organization has been possible on an urban scale.”

Proof? An archaeologist they draw on extensively for their account, John Chapman, indicates that the headcount Graeber and Wengrow invoke is based on a discredited “maximalist model.” Those thousand houses, he suspects, weren’t occupied at the same time. Drawing from at least nine lines of independent evidence, he concludes that these settlements weren’t anything like cities. In fact, he thinks a place like Taljanky may have been less a town than a festival site—less Birmingham than Burning Man.

A reader who does the armchair archaeology of digging through the endnotes will repeatedly encounter this sort of discordance between what the book says and what its sources say. Was Mohenjo Daro—a settlement, dating to around 2600 BC, on one side of the Indus River in Pakistan’s Sindh province—free of hierarchy and administration? “Over time, experts have largely come to agree that there’s no evidence for priest-kings, warrior nobility, or anything like what we would recognize as a ‘state’ in the urban civilization of the Indus valley,” Graeber and Wengrow write, and they cite research by the archaeologist Jonathan Mark Kenoyer. But Kenoyer has concluded that Mohenjo Daro was likely governed as a city-state; he notes, for instance, that seals with a unicorn motif are found throughout Indus settlements and infers that they may have been used by officials “who were responsible to reinforce the economic, political and ideological aspects of the Indus ruling elite.” Why should we hesitate to dignify (or denigrate) such a place with the name “state”?

Then there’s Mashkan-shapir in Iraq, which flourished four thousand years ago. “Intensive archaeological survey,” we’re told, “revealed a strikingly even distribution of wealth” and “no obvious center of commercial or political power.” Here they’re summarizing an article by the archaeologists who excavated the site—an article that actually refers to disparities of household wealth and a “walled-off enclosure in the west, which we believe was an administrative center,” and, the archaeologists think, may have had an administrative function similar to that of palaces elsewhere. The article says that Mashkan-shapir’s commercial and administrative centers were separate; when Graeber and Wengrow present this as the claim that it may have lacked any commercial or political center, it’s as if a hairbrush has been tugged through tangled evidence to make it align with their thesis.

They spend much time on Çatalhöyük, an ancient Anatolian city, or proto-city, that was first settled around nine thousand years ago. They claim that the archaeological record yields no evidence that the place had any central authority but ample evidence that the role of women was recognized and honored. The fact that more figurines have been found representing women than men signals, they venture, “a new awareness of women’s status, which was surely based on their concrete achievements in binding together these new forms of society.” What they don’t say is that the vast majority of the figurines are of animals, including sheep, cattle, and pigs; it’s possible to be less sanguine, then, about whether female figurines establish female empowerment. You may still find yourself persuaded that a preponderance of nude women among depictions of gendered human bodies is, as Graeber and Wengrow think, evidence for a gynocentric society. Just be prepared to be flexible: when they discuss the Bronze Age culture of Minoan Crete, the fact that only males are depicted in the nude will be taken as evidence for a gynocentric society. Then there’s the fact that 95 percent of Çatalhöyük hasn’t even been excavated; any sweeping claim about its social structure is bound to be a hostage to the fortunes of the dig.

And so it goes, as we hopscotch our way around the planet. If, a generation ago, an art historian proposed that Teotihuacan was a “utopian experiment in urban life,” we will not hear much about the murals mulled over and arguments advanced by all the archaeologists who have since drawn rather different conclusions. The vista we’re offered is exhilarating, but as evidence it gains clarity through filtration. Two half-truths, alas, do not make a truth, and neither do a thousand.

David Graeber

Peter Marshall/Alamy Live News

David Graeber, London, 2016

Of man’s first obedience: The Dawn of History has much of interest to say about the nature of the state. But if you take it to be a stress test of mainstream prehistory, you’ll discover that its aims and its deliverances are not quite in alignment. Indeed, when the dust, or the darts, have settled, we find that Graeber and Wengrow have no major quarrel with the “standard historical meta-narrative,” at least in its more cautious iterations. “There are, certainly, tendencies in history,” they concede, and the more reputable versions of the standard account concern not inexorable rules but, precisely, tendencies: one development creates conditions that are propitious for another. After agriculture came denser settlements, cities, governments. “Over the long term,” they grant, “ours is a species that has become enslaved to its crops: wheat, rice, millet and corn feed the world, and it’s hard to envisage modern life without them.” They don’t dispute that forager societies—with fascinating exceptions—tend to have less capital accumulation and inequality than sedentary farming ones. They emphatically agree with many evolutionists of the past couple of centuries that “something did go terribly wrong in human history.”

At the same time, a wholly plausible insight flows through The Dawn of Everything. Human beings are riven with both royalist and regicidal impulses; we’re prone to erect hierarchies and prone to topple them. We can be deeply cruel and deeply caring. “The basic principles of anarchism—self-organization, voluntary association, mutual aid—referred to forms of human behavior they assumed to have been around about as long as humanity,” Graeber wrote about the nineteenth-century anarchist thinkers in his Fragments of an Anarchist Anthropology (2004). “The same goes for the rejection of the state and of all forms of structural violence, inequality, or domination.”

The Dawn of Everything can be read as an effort to build out the “as long as humanity” thesis. We should readily accept that human beings routinely resist being dominated, even if they routinely seek to dominate; that self-organization, voluntary association, and mutual aid are vital forces in our social history. It’s just that Graeber and Wengrow aren’t content to make those points: they want to establish the existence of large, dense, city-like settlements free of rulers or rules; and, when the fumes of conjecture drift away, we are left without a single unambiguous example.

Whatever its empirical shortcomings, the book must be counted an imaginative success. Marx’s Capital came with an edifice of prehistorical and historical conjecture; the core tenets of Marxism do not stand or fall with it. The Dawn of Everything, too, has an argument to make that is independent of all the potsherds and all the field notes. In an era when social critique largely proceeds in the name of equality, it argues, in the mode of buoyant social prophesy, that our primary concern should, instead, be freedom, which it encapsulates as the freedom “to move, to disobey, to rearrange social ties.”

It must be said that Graeber, in books like The Utopia of Rules (2015) and Bullshit Jobs (2018), was exactly such a social prophet: satiric, antic, enthralling. It is impossible not to mourn the loss of that voice. His vision is of particular significance because it doesn’t fit easily within the usual political positions of our era. What readers of The Dawn of Everything should not overlook is that the sort of inequality we mainly fret about today is of concern to its authors only inasmuch as it clashes with freedom. At its core is a fascinating proposal about human values, about the nature of a good and just existence. And so we can profitably approach this book with Rousseau’s disclaimer in mind: “One must not take the kind of research which we enter into as the pursuit of truths of history, but solely as hypothetical and conditional reasonings, better fitted to clarify the nature of things than to expose their actual origin.”

Yes, plenty of arguments can be made against Graeber and Wengrow’s anarchist vision; some double as arguments against libertarian ones. (Note, for instance, the paradoxical nature of the “freedom to disobey”: we cannot be commanded—and therefore we cannot disobey commands—without institutions that authorize command.) But these are, precisely, arguments; they could be wrong, in part or whole. They should be weighed, assessed, tested, and perhaps modified in the face of counterarguments. And the worst argument to make against anarchism—against a polity without politics—is that we haven’t quite seen it up and running yet. “If anarchist theory and practice cannot keep pace with—let alone go beyond—historic changes that have altered the entire social, cultural, and moral landscape,” the eminent anarchist Murray Bookchin wrote three decades ago, “the entire movement will indeed become what Theodor Adorno called it—a ‘ghost.’” We live in an era of the World Wide Web, same-sex marriage, artificial intelligence, a climate crisis. We don’t need to peer into our prehistoric past to decide what to think about these things.

“Pray, Mr. MacQuedy, how is it that all gentlemen of your nation begin everything they write with the ‘infancy of society?’” an epicurean reverend asks a political economist in Thomas Love Peacock’s novel Crotchet Castle (1831). This habit, so entrenched in that era, has persisted through ours. The Dawn of Everything sometimes put me in mind of Riane Eisler’s international best seller from 1987, The Chalice and the Blade. Eisler, trawling through the Neolithic, saw a once-prevalent woman-friendly “partnership” model of “gylany” being supplanted, in stages, by the “dominator” model of “androcracy.” Like Graeber and Wengrow, she had a deep antipathy toward domination; like them, she cherished a vision of freedom and mutual care; like them, she thought she glimpsed it in Minoan Crete.

But the moral argument here doesn’t depend on whether we believe that gylany was once widespread: an ancient pedigree doesn’t make patriarchy right. Social prophets, including those in the anarchist tradition—from Peter Kropotkin and Emma Goldman to Paul Goodman and David Graeber—make the vital contribution of stretching our social and political imagination. Facing forward, we can conduct our own experiments in living. We can devise the stages we’d like to see.

That’s what Rousseau came to think. By the time he published The Social Contract (1762), he had given up the notion that political argument needed to be buttressed by some primordial utopia. “Far from thinking that neither virtue nor happiness is available to us,” he argued, “let’s work to draw from evil the very remedy that would cure it”—let’s reorganize society, that is, through a better social compact. Never mind the dawn, he was urging: we will not find our future in our past.

Owing to an editorial error, an earlier version of this article did not include the author’s final changes. The text has been corrected to reflect them.